The Art of Data Communication: Simplify, Visualize, and Storytell
I am an engineer and a data leader from India, living in the United States. English being my second language, I am hyper-aware of how I communicate. Upon arriving in the United States, I was struck by the efficiency and clarity of native English speakers. This sense of awareness makes me think about the importance of effective communication. As a data leader, my team and I may have the best tech and data solutions, but communicating them and getting the business partners and clients on board is a critical part of my job. Without their buy-in, I might as well be just building toys in my garage and patting myself on the back!
Storytelling is a key part of being a data professional. These are some of the things I have learned from reading books and articles on storytelling in cinema, as well as in the context of data communication -
1. Have goals and a clear message
Before you start building models or analyzing data, try to have clear goals of what you are aiming for. Unless it’s an exploratory analysis, you should be able to define goals ahead of time.
2. Keep it simple
With any experience in data and AI, you quickly come to the realization that there just are too many problems to solve, and each problem can have multiple solutions. We have a saying at the place I work — Give me the story, and I can find the data to support it ;) Going back to the previous point, having a clear goal helps, and so does keeping it simple and iterating through the problems.
3. Know your audience
Depending on who you are talking to and how fluent they are in data, you will need to tailor your presentation or talking points. For example, an executive might be less interested in how you arrived at the solution but more focused on your work's value to the organization. Whereas a presentation to the engineering team needs to have technical details of your solutions or models.
4. Set the right expectations
I have come across executives who believe data and AI models will solve the world’s problems and those who would rather go off their gut/wisdom and just use data to further their beliefs. It is important to set the right expectations on what you can deliver and communicate that data science is an iterative process.
5. BI and visualization are your friends
BI (Business Intelligence) and visualization tools can be extremely useful when presenting data insights. Good visualizations can drive home the point you are making without much effort. Some executives are numbers people — so again, knowing your audience and their preferences helps.
6. Create a story arc
People love stories; they are a powerful tool for communicating actionable intelligence. Speak from the point of view of your audience — they are the protagonist of the story, and you are a helpful guide or advisor helping them achieve their goal.
7. Be honest about assumptions, challenges, and limitations
In my experience, it’s just better to be open and honest about your work — the limitation and doubts you may have with your models. It helps build trust and credibility. It’s not always easy, but it works for the better over the long run.
8. Add humor
Data and analysis can be dry, but adding humor can help to lighten the mood and make your presentation more engaging. The good news is, you don’t have to be good at it. The bar humor in the corporate world is so low that even trying out mundane (appropriate) jokes with a meh-delivery gets you street cred!
Conclusion
Effective communication is a vital aspect of being a successful data professional, and aspects of storytelling can be a powerful tool. I have learned that there is always room for improvement and that being open to feedback and continuous learning is crucial for growth in this field.
